What is a streaming delay?

To be able to understand the delay (also known as latency) in streaming, it is important to understand the difference between video conferencing and streaming. 

Video conferencing is a two-way collaboration between a restricted number of people (using tools such as Skype, FaceTime, etc.).

Streaming is when there are a number of presenters that speak to a large audience in multiple locations, in a one-way collaboration.

When streaming a presentation to multiple locations, the video is being compressed, transferred and received at all these locations using the available bandwidth.

The accumulation of these actions results in a latency of 30 to 35 seconds, depending on the factors mentioned. This will mean that the presentation in real time will be streamed with a slight delay; at SpotMe an average of 30 seconds is expected on our platforms.

Managing delays during interactivity

In order to get the best out of interactivity while taking latency into account, we suggest the following best practices with our standard interactive tools:

Live polling

Scenario 1

When a polling question is asked to the audience and they begin to answer, it is important to leave enough time (for example 45 seconds) for the question to be answered by the audience. This will leave sufficient time for the question to be live on screen while the presenter is discussing, while also ensuring that the latency is taken into account.

Scenario 2

A polling question could also be asked at the beginning of the session and left live throughout the duration of the session. This leaves the question open for the audience to submit their answers and change them throughout the session as the presentation develops. The presenter will be able to analyze and comment on the results as they change throughout the whole session.

Q&A 

Q&A questions can be submitted throughout the session in real time and asked whenever needed by a moderator to the speaker. 

If questions are displayed live on the stream, it is important for the host to read the question out loud before answering it, in order to not have any confusion when the question is answered due to the delay of the question appearing.